The "Captain Jamil Hussein" problem, an incident in which the Associated Press appears to have used a fake policeman as a source for an extended period of time, raises the generic problem of how to verify news stories generated by a field correspondent. The problem is not confined to the MSM. Bloggers are increasingly bombarded with tips and reports from individuals they may have corresponded with, but whom they have never met, and whose information may be questionable. There are great opportunities under those circumstances for pranksters and malevolent persons to spread disinformation. Although the problem is most marked in the case of the mainstream media (MSM) it is not confined to them. The problem of acquiring accurate intelligence, even in the face of active disinformation, is a challenge to everyone. Hence, the "Captain Jamil Hussein" problem is typical of a class of challenges probably best understood in the context of a simplified intelligence cycle: that of collection-analysis-dissemination.
The press, because of its huge institutional presence is still the primary open source mechanism for "collection". The blogosphere has increasingly become a major player in "analysis". As the blogosphere continues to expand and MSM media personalities start to blog, the division of labor between these two entitites will start to blur. Organizations like Pajamas Media, Oh My News! and 18 Doughty Street are starting to provide original reportage -- collection -- while continuing to focus on analysis. But however the labor is divided, the conceptual distinction between collection and analysis remains and we will return to the subject later.
Dissemination, that is, the way particular stories are chosen from among the many thousands of daily events and processed between Internet centers of consciousness until they are finally served up to their relevant audiences -- is of crucial importance too. This describes the path that a "story" takes from its inception as it is added to, modified, and amplified until it becomes a subject of wide public interest. The classic example of this was the Dan Rather Killian fake document event. CBS itself collected the information by posting the faked documents. But it was analysis, beginning with the Free Republic and spreading through the blogosphere that was truly revolutionary. The detection of the fake, first expressed as a vague doubt, then focusing on increasingly specific and incontrovertible technical criticism, was amplified as it was picked up by progressively "larger" centers of consciousness -- larger in the sense of audience reach and prestige -- until it amounted to a tidal wave that discredited Sixty Minutes, forced Dan Rather into retirement and cast several major media personalities into the outer dark.
But as the MSM itself was fond of pointing out, collection remained largely their monopoly at least for "general interest news". The event horizon for the blogosphere still remains the wire service release or dispatch filed by newspaper or media correspondents, which however clever bloggers might be, they could not peer past. And it was in this shadow zone where the "Jamil Hussein" problem took place. The MSM dominance in specialist news is far less complete. Global Voices, a weblog community run by the Harvard Law School, has systematically been hosting or linking to weblogs authored by its contacts in largely obscure Third World locations. Global Voices is often the only primary source of events in places where the MSM has no presence because lack of interest makes it impractical to station correspondents there; and on which even editorial time is deemed wasted unless some war or natural disaster should strike. For many analytical bloggers, Global Voices (and others like it) is the event horizon for "under the horizon" or chronic events in obscure places like Timor or Chad. The trade press often provides the first reportage on subjects too obscure or uninteresting to the general public for WaPo of the NYT to notice. Sites like the Fourth Rail or MEMRI often have private sources which regularly scoop the MSM on subjects like events in Pakistani tribal areas or trends in Arabic news stories. Yet while the MSM doesn't provide private collection in these niche areas, it is still true that it owns the publics first glimpse of the general news.
Bearing that relationship of the MSM and blogosphere in mind, one may ask, what are the desirable properties an interface that bloggers could use to fix their greatest weaknesses? The previous discussion identifies two major issues:
- Pushing the event horizon back, beyond that provided by the MSM wire service system; and facilitating the verification of reported stories; and
- Making it easier for bloggers to "disseminate" information. That is to make their information available in such a way that it can be more easily picked up and amplified until it reaches a wide audience.
The first thing to observe is that once information is entered on the Internet, it is potentially "discoverable". RSS, for example, has a created a standard by which any content publishes the equivalent of a table of contents and makes it possible for software to "discover" what it is about. The ideal blogging interface can build on this by incorporating "automatic discovery" into its design. At any given time there are a number of conversations on the Internet on subjects which are potentially knowable, provided they are formatted according to an agreed standard which states their content. If a blogger could automatically be cued into ongoing or past threads as he is posting on a subject or immediately thereafter, it becomes potentially possible to "join" that thread, even though the thread only exists conceptually. For example, it would probably be useful if a medical researcher, posting on his latest research findings automatically became aware of researchers engaged in similar or related pursuits. To use another example, it would have been very useful if a blogger when writing about "Captain Jamil Hussein" became conscious of the many threads in which the shadowy Captain featured. All of these things can be done with current tools, using search engines, RSS aggregators and the like. However they take time and effort. Time and effort which only a few bloggers like Flopping Aces seemed ready to invest. The AP either did not know how or were too busy to make undertake an equivalent task. Many non-Internet savvy users won't know how to what Flopping Aces did in any case. However, if these features were built into a blogging interface itself, preferrably as an underlying standard which could be implemented creatively by developers, then this power would be at the fingertips of laymen.
The ideal blogging interface should also be able to initiate and accept messaging between other related threads, whether or not those threads were generated by human authors or programmatically. For example, it would be desirable for someone posting on the subject of "Captain Jamil Hussein" to queue messages to other threads without having any special knowledge about the identities of the authors of those threads. This will make it easier for relavant information on a subject to "snowball" in such a way that the knowledge base on the subject is expanded while simultaneously signaling the state of development of the story. This will facilitate the amplification and transformation of a meme and shorten the period between the time an event leaves the news event horizon and becomes a subject of wide public interest. Again these things can be done with current tools, such as search engines, email, instant messaging, telephony and whatnot. But inconvenience and lack of knowledge among users will limit the process of dissemination unless these features can be built into the interface to make it easier.
What will this achieve? First of all, it will transform the blog, which is an ill-defined piece of generic information into a powerful object with its own properties and methods. This will make each post "live". It will be alive to the downstream and upstream; that is to the collection and dissemination. But more importantly it will create the architecture to make blogs part of the collection system itself. This will make it potentially possible for a blogger to file something, perhaps even unintentionally, about "Captain Jamil Hussein" or any other subject and add to a conversation. From the point of view of the collection-analysis-dissemination cycle, it will push back the event horizon, provide collateral sources of confirmation to existing ones and provide better ways to disseminate information outwards.
There are many technical issues which I have omitted to mention. Permissions, for example. Ideally authors should be able to choose the degree of discovery they will allow on their material. Blog authors may wish not only to control the outgoing information by limit by filters the related discoveries they are interested in seeing. There is the issue of whether the underlying web services to support discovery and messaging can be feasibly constructed and paid for. There are issues of latency. But whether or not these specific ideas make sense in and of themselves, the challenges facing the open source intelligence system today are formidable and require architectural solutions. Talent and luck will still play a part. But design will play a bigger one.